Here is how to "tweak" x265 like master $@3&@

 

The program is so insecure it is using the windows clock to calulate your score.

 

You simply start the bench and change the windows time back 1 minute for however long you want, longer for better score on high core count chips then change it back before the bench is done.

 

 

 

This is not a glitch, this is not a bug. This IS a cheat. You can check a legit run side by side to a cheat run. The timer in the bench

 

will show much longer than normal by 15-20% IE 12 seconds instead of 10 seconds. The fact that he didnt use overkill is what made me

 

realise immediately this is a cheat run. Here is an example of overkill 2x and no overkill. I studied this bench for 2 days straight

 

trying all sorts of BS tweaks that are suggested and everyone that helped (thankyou) came to the same conclusion its impossible for no

 

overkill to beat 2x under and circumstances.

 

Here is the 10 core WR on air

 

Global WR on air with 10 core air

 

Validates not problem

 

 

I feel totally dumbfounded by this. If an angry gorilla in his basement can figure this out then what else can be manipulated??? Well I'll

 

tell you what. Hwbotprime.

 

Here is 2nd place at 4.5ghz got tired of feeling dirty trying for first. Same method used. Change 1 minute backwards then 1 minute forwards before the end.

 

It seems people that know this think its funny to suggest that you need to test all the version of java to find

 

the best. If you are one of those people you can rot in your own special form of hell. Ive spent countless time benching hwbotprime and

 

testing every little nuance and to see this makes me physically ill. So we can guess who has first place in that benchmark also $@3&@....

 

I call for a ban $@3&@ (1 month leave his scores up and give him two more chances lmfao this place is honestly turning into a joke) and removal of points for hwbotprime and x265 till further notice. Anything less will feel extremely disappointing. Feeling totally unmotivated now, and I feel like nothing will even be addressed so carry on as usual

Edited August 23, 20178 yr by Splave

Great find man. Both benches have to be removed/disabled points. There is no other way in my eyes.

What OS?

 

Just tried Windows 10 with HWBOT Prime and changing the Windows time just pauses the benchmark.

Ugggh...looks like an obvious cheat... points definitely need removal...

What OS?

 

Just tried Windows 10 with HWBOT Prime and changing the Windows time just pauses the benchmark.

 

Says windows 7 in the benchmark slacka..

/me grab popcorn...

 

This sucks tho...

I noticed a lot of weird stuff whilst looking at his submissions, nice to know you got to the bottom of it. Thanks Splave.

Ridiculous... I've been testing almost all Java versions and so on!!! Remove points!

 

Gesendet von meinem ONEPLUS A5000 mit Tapatalk

What OS?

 

Just tried Windows 10 with HWBOT Prime and changing the Windows time just pauses the benchmark.

 

NVM, that's expected behaviour.

 

Re-created on HWBOT Prime, now testing X265 benchmark.

 

@havli @richba5tard

What OS?

 

Just tried Windows 10 with HWBOT Prime and changing the Windows time just pauses the benchmark.

 

Need to start the bench flip back a minute, then it will "pause" but not really paused just looks paused. Then flip minute forward. Profit

 

Whoever coded these should an hero.

Need to start the bench flip back a minute, then it will "pause" but not really paused just looks paused. Then flip minute forward. Profit

 

Flipping forward isn't necessary, at least not for HWBOT Prime. The benchmark just halts and picks up after a while.

 

Do you have the same behavior with older Java versions?

needs points removal, period. atleast till fixed!

Im using 113 and 107

Yeah would be wise to halt points for these two until new versions are released.

So sad to see.

 

Great work Splave

Im using 113 and 107

Thanks for your tweaks. /s

Im using 113 and 107

 

OK, thanks.

 

Are you able to control the scoring? So far I'm seeing large differences in scores when changing only 1 minute (2.5x higher).

 

Do you see negative scoring in X265?

 

 

The feels right nao. . .

 

@PJ you are doing it wrong if the scores negative. Just need to hop in to the Delorean and apply 1.21 jiggawatts.

Edited August 23, 20178 yr by l0ud_sil3nc3

@PJ you are doing it wrong if the scores negative. Just need to hop in to the Delorean and apply 1.21 jiggawatts .

That made my day man!!!

 

Gesendet von meinem ONEPLUS A5000 mit Tapatalk

OK, thanks.

 

Are you able to control the scoring? So far I'm seeing large differences in scores when changing only 1 minute (2.5x higher).

 

Do you see negative scoring in X265?

 

 

 

You need to start go back one minute in time wait about 3 second then put back forward 1 minute. Bench will jump to halfway done. And score will be high.

 

 

Regardless you are changing the score by changing the clock. Not good.

You need to start go back one minute in time wait about 3 second then put back forward 1 minute. Bench will jump to halfway done. And score will be high.

 

Yeah, I had some positive scores as well. But all pretty much 2.5X higher than normal.

Yeah would be wise to halt points for these two until new versions are released.

 

 

Maybe you can trade him an xtu valid file for one of his x265? Too soon?

Oh great i read all of this to lern to be a l33t bencher ..